You read my mind! As I said before I am sorry to honest but yes you are exactly right that Qlikview isn't designed to make such comparisons. I am familiar of JAVA and SQL Server and this tool is totally new thing for me. Like JAVA or any other programming language, Qlikview cannot be bent as and how you want sadly. I am currently on the job training for Qlikview only and Qlikview 11 for Developers has really helped me know how the tool works in a very short time. I really like the associativity power of this tool to make very fast analysis unlike other tools which use a drill down approach but sadly I was accustomed to programming in Java and likewise and at times I feel weak when I cannot accomplish things in a simple way. And another thing is finding a proper solution is difficult for this tool since there are not so many people as there are on Stack Overflow which I used to solve any programming doubts. But thanks for making it clear to me. As a fresher in this tool, I want to move fast but I guess I'll have to rattle the cage of my mind to find those answers or leave those questions unanswered. Again sorry if was being harsh or rude
I am great admirer of your solutions in Qlikview and QlikSense.
I also have some similar requirements where I need to build a solution of data comparison from 2 different systems. Its basically a data migration from 2 systems. The data from system A getting updated in System B. The extract from System A extract will be delta on daily basis and need to be compared with System B data which need to be in Sync. There are chances that multiple fields may have wrongly updated or missed during the course of migration to system B. I need to develop an app which compare the data mismatch from Stem B wrt to system A. There may be chances that 100 records sent from Sys A and only 70 records getting updated. And next day after the data fix, there may 2nd day data plus the fixed data gets updated in System B.
Can you please suggest the best approach which can help me getting this developed in QlikSense. I believe, if its the same number of records and same records gets into both system with unique transaction id, we can compare one to one in same row. But id there is missing records or more records, I doubt over the solution.
Well, the solution here does highlight rows missing from one or the other table, but I don't know if this sort of script and table works in QlikSense. If not, I won't be much help, as I've not played around with QlikSense.
The best way to do this would be to load the data from both the sources in to one application (QVF/QVW) and have a field to differentiate data from both sources. In this way you will be able to leverage the power of the Qlik associative engine and also find the missing information you are looking for simultaneously.
I need same, but using NOT two separate files, but merged in one cube using load with * : C:\Users\...\orders_*.xlsx
so how to mark PO numbers not matching = POs added in file in 0628?